Xinyi Li

@sdcb Hi, 请问你在进行推理的时候是否都是用的官方推理库？我们这边check了一下，发现官方的推理库都是用带avx512的机器生成的。在这种情况下，如果在非avx512机器上用官方推理库，可能就会导致这种情况，类似[#11597](https://github.com/PaddlePaddle/PaddleOCR/issues/11597)。针对这个issue我们已经提了[PR](https://github.com/PaddlePaddle/Paddle/pull/64132/)，等merge之后应该就能解决了。

Hi, this issue should be fixed on **develop branch** of Paddle after oneDNN **v3.4** is merged. The root cause is, some specific kernels (brgemm kernels) in oneDNN were not reused...

> @LLee233 Thanks for mentioning it, can you mention the commit for oneDNN v3.4. I can't see any changes regarding oneDNN v3.4 on any branch. @Shubham654 [This commit](https://github.com/PaddlePaddle/Paddle/commit/f654bc16be3dbfcbc9cf30ea147428bbb7687010) is where...

@onecatcn Hi, would you mind helping find approvers as below? Since this PR includes OP API changes:  Thank you!

> Hi, about ConvBiasFusePass, I have some questions. In previous pass( non-pir implementation), I run ResNet50, and found it not fused in conv_bias pass, but fused in conv_elementwise_add pass. But...